The choice of reason isn't arbitrary. You could say that they have stripes because it helps them shoo away flies and someone else could say they have stripes because it makes them look pretty - but one (or both) of these answers would be wrong.
The hypothesis can be falsified. We could test whether zebras do find stripes pretty in other zebras - perhaps it's not the best way to spend your time and resources, but it can be done. We can measure whether stripes discourage flies from landing, and we can measure what impact (if any) less flies has on health of zebras and their likelihood to survive until breeding. It's tricky, but it's generally possible.
There's some mechanism that makes stripes beneficial. Or perhaps there are multiple valid mechanisms - but in that case, one contributes (for example) 70% and the other 30%, and that can (with enough effort) be evaluated, and those mechanisms separated from the other alternative answers that are just guesses.
There are definitive answers. Perhaps they're hard to find out, perhaps they're hard to verify, perhaps for some aspects we'll not find out these definitive answers, but they certainly are there. You may reasonably choose that a particular answer isn't important enough to justify the resources needed to obtain that answer (and zebra stripes may well be a decent example for that), but giving in to the concept "there's really no definitive answer, except whichever one you choose to accept" is just shameful wilful ignorance. Reality is what exists no matter if you choose to accept it or not.